OpenAI · Thursday, September 24, 2026
OpenAI’s Sam Altman tells U.N. Security Council AI must stay under human control
The File desk · Sep 24, 2026, 10:23 AM UTC
Status
Confirmed (OpenAI remarks post, Sept. 23, 2026)
On Sept. 23, 2026, OpenAI posted remarks Sam Altman delivered to the United Nations Security Council. He said the company should not train models it cannot make an extremely strong case it will keep under human control. The post names two failure modes — loss of control, and concentration of power — and calls for shared frontier standards, incident reporting, and a secure channel for vulnerabilities.
On Sept. 23, 2026, OpenAI posted remarks its chief executive, Sam Altman, delivered to the United Nations Security Council.
He said: “we should not train models that we cannot make an extremely strong case that we will be able to keep under human control.”
The post describes compressed timelines and two failure modes: loss of control, and concentration of power.
He urged complementary national and international standards for frontier artificial intelligence, incident reporting, and a secure channel for vulnerabilities.
The desk has OpenAI’s post of the remarks. A United Nations video or transcript beyond that post was not fetched.
What is still unknown or disputed
- United Nations primary video or transcript beyond the OpenAI post.
Primary sources
Every claim in this story is drawn from the documents below. If a fetch failed, that is recorded on the card.
Source 1
Sam Altman addresses the U.N. Security Council
OpenAI · September 23, 2026
Today, OpenAI CEO Sam Altman addressed the United Nations Security Council on artificial intelligence… we should not train models that we cannot make an extremely strong case that we will be able to keep under human control.
https://openai.com/index/sam-altman-un-security-council-remarks/